Wednesday, January 9, 2008

Is Jeff Sachs wrong again?

After his shock therapy policy went awry in Russia, Jeff Sachs' new credo is to solve World poverty, and specifically African poverty, by eliminating malaria. While it is clear that it is a very serious disease, one may ask why so much relies on this one disease.

Jeff Sachs offers the answer himself in several papers, the most prominent ultimately published in The American Journal of Tropical Medicine & Hygiene (pdf) (one can wonder whether this paper would have passed peer review in an Economics journal): in terms of growth or GDP, his cross-country regressions show a very substantial cost, in the order of 1.3% growth points or 40% of the GDP level. His latest work corroborates what he has saying alongside Bono and Angelina Jolie on MTV for a long time: controlling malaria is cheap, about US$4 a person at risk.

The typical prevention that is advocated is the bed net that is treated with a substance that disrupts the cycle of the disease. Bed nets are relatively cheap and effective, as long as you stay under them. Jeff Sachs' goal is to get everyone under a bed net, and he has drummed up public support for bed net distribution campaigns in several countries. This initiative has recently come under fire, as discussed in this Science article: these distributions of free imported bed nets kill nascent local bed net industries, and once the campaigns are over, no one is there to take over. This also highlights the long standing issue of aid and dependence.

A second issue is that Sachs presumes that malaria is exogenous to the level of development. What if it is endogenous to poverty? Then, one should deal with poverty first to get rid of malaria, and not deal with malaria to get rid of poverty. The truth is probably that they are both endogenous to each other. Encouraging a local market for bed nets to thrive, instead of carpet bombing it with bed nets, is probably the better way to go.


Anonymous said...

Jeff Sachs used to be a good economist. Now he is just going for the money and the photo-ops with celebrities. Pretty sad, and the press is falling for it.

Anonymous said...

Does anybody still believe in cross-country regressions? How people can rely on the results of regressions with tiny samples, huge measurement errors and gigantic endogeneity problems baffles me. And one can wonder, in this case, how Jeff Sachs came up with reliable malaria data.

Anonymous said...

I found this interesting paper, that uses theory, properly calibrated, to evaluate the the cost of malaria. It uses theory to bypass the endogeneity problem I mentioned. One interesting feature is that it allows people to buy protection, if they can afford it. It turns out they will, as long as protection is effective. The cost of malaria is negligible then. But even minor ineffectiveness would lead to very significant costs, in the order of what Sachs seems to indicate.

Bruce Webb said...

Well as to where the data came from the following would seem to be a solid possibility.

Why malaria? Well it seems to be the treatable disease that costs most deaths. In Angola you are three times more likely to die of malaria than measles.

As to the question of endogenous or not to poverty the answer seems pretty clear. Poor people would not seem to intrinsically create heat and stagnant water which is pretty much as you need to breed mosquitos to transmit the malaria parasite.

Plus the nets are not just nets, they are treated with DDT. I am not sure it makes a lot of sense to set up hundreds of local factories to apply DDT to nets, I think you would want to maintain a little more control. And the reason they work is not because people huddle under them all day, by and large the mosquitoes are nocturnal and more likely to get a full meal when the victim is sleeping.

Anonymous said...

Malaria data may be available, but its reliability should be highly questioned. I do not trust any of Jeff Sachs' results on this.
If he manages to get attention on Africa with this work, fine. But I would prefer he were to manage that with believable results